Don't re-run job until previous prologs done
If a job was requeued while some prolog was running (e.g. waiting for node boot, burst buffer, PrologSlurmctld, etc.) then wait until that completes before considering starting the job again.
Please register or sign in to comment